How the Event Risk Calculator - Coronavirus Works
This tool estimates the probability that at least one person at a gathering is currently infectious with COVID-19, using the same underlying logic as the Georgia Tech COVID-19 Event Risk Assessment Planning Tool (Chande et al., "Real-time, interactive website for US-county-level COVID-19 event risk assessment," Nature Human Behaviour, 2020). It combines your event size with a local, per-capita estimate of active infections derived from recently reported cases and an adjustable undercounting multiplier.
The formula
First, the calculator estimates the chance that any single random person in the area is currently infectious: p = (ascertainment bias × reported cases) ÷ population. Then it applies the complement rule for independent events to estimate the chance that at least one of your n attendees is infected: Risk = 1 − (1 − p)^n. For example, with 750 reported cases over 10 days, a 5x undercount multiplier, and a population of 500,000, p ≈ 0.75%; for a 50-person event, Risk = 1 − (1 − 0.0075)^50 ≈ 31%.
Why an undercount multiplier?
Officially reported case counts miss infections that are asymptomatic, never tested, or tested at home without being reported to public health authorities. During the pandemic, researchers and health departments commonly estimated true infections at roughly 3 to 10 times the confirmed case count, depending on testing availability and reporting practices at the time. Raising the multiplier raises the estimated risk; lowering it (toward 1x) assumes reporting was close to complete.
What this calculator does not model
The result is a presence-of-infection estimate, not a transmission-risk estimate. It does not account for masking, ventilation, indoor versus outdoor settings, vaccination status, prior infection, test-to-enter policies, or the transmissibility of a specific variant — all of which materially change the actual chance of transmission at an event. It also assumes cases are evenly distributed across the population rather than clustered. Treat the output as one planning input, not a certification that an event is "safe," and consult current public health guidance and, for personal medical questions, a healthcare provider.